Selectivity control towards CO versus H2 for photo-driven CO2 reduction with a novel Co(II) catalyst

Developing efficient catalysts for reducing carbon dioxide, a highly stable combustion waste product, is a relevant task to lower the atmospheric concentration of this greenhouse gas by upcycling. Selectivity towards CO2-reduction products is highly desirable, although it can be challenging to achieve since the metal-hydrides formation is sometimes favored and leads to H2 evolution. In this work, we designed a cobalt-based catalyst, and we present herein its physicochemical properties. Moreover, we tailored a fully earth-abundant photocatalytic system to achieve specifically CO2 reduction, optimizing efficiency and selectivity. By changing the conditions, we enhanced the turnover number (TON) of CO production from only 0.5 to more than 60 and the selectivity from 6% to 97% after four hours of irradiation at 420 nm. Further efficiency enhancement was achieved by adding 1,1,1,3,3,3-hexafluoropropan-2-ol, producing CO with a TON up to 230, although at the expense of selectivity (54%).


General information
The starting materials, solvents, and reagents were purchased from commercial suppliers and used without further purification unless stated.Anhydrous N,Ndimethylacetamide (<0.005% water) was purchased over molecular sieves, while triethylamine (TEA) and triethanolamine (TEOA) were purified by distillation.For the photocatalytic tests, 20 mL-glass vials were purchased from TH.Geyer, with a septum.NMR spectra were recorded on a Bruker Advance 400 NMR instrument at 400 MHz for 1 H NMR and 126 MHz for 13 C NMR.The NMR spectra were recorded at 22 °C in deuterated solvents.Electron spray ionization (ESI) experiments were recorded on a Thermo Fisher Scientific Q-Exactive (Orbitrap) mass spectrometer, at the Institute of Organic Chemistry (Karlsruhe Institute of Technology).

Syntheses Photosensitizer [Cu(dmp)DPEPhos](BF4)
The synthesis of the heteroleptic Cu(I) complex used as photosensitizer was performed according to the general procedure, already published [1].In a Schlenk tube, under an inert atmosphere (Ar) a solution of DPEPhos and Cu(CH3CN)4BF4 in dry CH2Cl2 was stirred for 0.5 h at room temperature.One equivalent of 2,9-dimethyl-1,10-phenanthroline was added and the reaction mixture was stirred at room temperature for an additional 5 h.The solvent was removed under vacuum and the residue was purified by column chromatography on silica gel (eluent: CH2Cl2; Rf: 0.39) to give the desired compound as a yellow powder with 67% yield. 1

Ligand
The synthesis of the ligand for the catalyst was achieved in two steps following the experimental procedure published previously [3].Caution must be taken when handling organic azides as they might be explosive.

Cobalt catalyst, 1
In a two-necked round-bottomed flask, under argon, the chelating ligand BzQuTr (100 mg, 0.35 mmol, 2.0 equiv) dissolved in 10 mL of dry MeOH was added dropwise to a solution of Co(NCS)2(py)4 (86 mg, 0.175 mmol, 1.00 equiv) in 5 mL of MeOH.The mixture was stirred for two hours at room temperature.The solvent was removed under reduced pressure and the crude product was washed with cold MeOH and Et2O, obtaining a lilac precipitate (82 mg, 0.11 mmol, 60%

X-ray analysis, experimental details
Single crystal X-ray diffraction data were collected on a STADI VARI diffractometer with monochromated Ga K (1.34143 Å) radiation at low temperature (150 K for 1a and 180 K for 1b).Using Olex2 [4], the structures were solved with the ShelXT [5] structure solution program using Intrinsic Phasing and refined with the ShelXL [6] refinement package using least squares minimization.Refinement was performed with anisotropic temperature factors for all non-hydrogen atoms; hydrogen atoms were calculated on idealized positions.Crystallographic and structure refinement data of the two polymorphs of 1 (1a and 1b) are summarized in Table S1.
Crystallographic data for compounds 1a and 1b reported in this paper have been deposited with the Cambridge Crystallographic Data Centre as supplementary information no.CCDC-2285968 and 2285969.Copies of the data can be obtained free of charge from https://www.ccdc.cam.ac.uk/structures.

Electrochemical measurements
Electrochemical experiments were recorded on a Gamry Interface 1010B potentiostat.
The electrochemical cell was equipped with three electrodes: a glassy-carbon disk as a working electrode, a Pt wire as an auxiliary electrode, and a Ag wire as a quasireference.Internal standards were necessary to calibrate the potential and ferrocene (Fc) or decamethylferrocene (Me10Fc) were used, according to the electrochemical window.The potentials of Fc and Me10Fc differ by about 450 mV in DMA with our S7 system.For comparison with the values, please refer to literature sources [7][8][9].The experiments were performed using tetrabutylammonium hexafluorophosphate as a supporting electrolyte (0.1 M).Scan rates were typically 100 mV/s.

Photo-driven CO2 reduction
Experiments were performed in a photoreactor from Luzchem (model: LZC-ICH2) equipped with four lamps at 420 nm and four mini-stirrers.On each stirrer, two samples were irradiated at the same time, for a total of eight simultaneous reactions.Typically, the solutions contained the photosensitizer (1 mM or 0.   S11

Figure S3 : 1 Figure S4 .
Figure S3: FTIR spectrum of complex 1, measured by attenuated total reflection (ATR), was recorded with a Bruker Alpha P instrument.

Figure S8 :
Figure S8: Cyclic voltammogram of ligand in DMA, 0.1 M TBAPF6, recorded at a scan rate of 100 mV/s.Potentials are reported versus the internal reference ferrocene.
5 mM), catalyst 1 (different concentrations studied), and BIH (usually 10 mM or 20 mM), unless otherwise noted.At the irradiation wavelength (420 nm) the optical density was 0.35.The temperature of the reactor was controlled with an in-built ventilator, T = (25 ± 5) °C.The moles of products (CO and H2) were measured by quantitative analyses of the headspace of the reactions with a gas chromatograph from Shimadzu (GC-2030) equipped with two barrier discharge ionization detectors (BID).Every test was repeated at least twice.

Figure S9 :S10Figure S10 :
Figure S9:(Left) Explanatory drawing of the set-up, observed from above.In the photoreactor chamber, the samples (yellow circles are put at a distance of 1 cm from the light source on a mini-stirrer (green rectangular shapes).On a mini-stirrer, two photocatalytic systems were put.(Right) A picture of some irradiated samples in the photoreactor, taken after photocatalysis.

Figure S11 :
Figure S11: Monitoring of the absorption changes during a typical photocatalysis, DMA/TEA 7:1 (v/v), PS: 1 mM; catalyst 0.01 mM; BIH 10 mM; under irradiation at 420 nm over the period of 4 h.(Asterisk shows the lamp change of the spectrometer).

Figure S12 :
Figure S12: Monitoring of the absorption changes of the catalyst (0.001 M) in a solution DMA/TEA 7:1 (v/v) during the first hundred minutes under irradiation at 420 nm, small decrease of the d-d band is visible, and a slightly increased absorption at around 400 nm can be detected.The solution was left in the dark for 4 days, and the evident spectral changes indicate the instability of complex 1 in the solution after this long timeframe.

Table S1 :
Crystallographic and structure refinement data of polymorphs of 1.

Table S2 :
Selected bond lengths and angles of polymorphs 1a and 1b.